116,336 Hits in 8.6 sec

A Multi-Objective Approach to Mitigate Negative Side Effects

Sandhya Saisubramanian, Ece Kamar, Shlomo Zilberstein
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Agents operating in unstructured environments often create negative side effects (NSE) that may not be easy to identify at design time.  ...  The slack denotes the maximum deviation from an optimal policy with respect to the agent's primary objective allowed in order to mitigate NSE as a secondary objective.  ...  Our proposed framework is shown to be effective in mitigating undesirable side effects.  ... 
doi:10.24963/ijcai.2020/50 dblp:conf/ijcai/SaisubramanianK20 fatcat:vj226fqd2vdxdo2iu4iep7jk6i

Avoiding Negative Side Effects due to Incomplete Knowledge of AI Systems [article]

Sandhya Saisubramanian and Shlomo Zilberstein and Ece Kamar
2021 arXiv   pre-print
This article provides a comprehensive overview of different forms of negative side effects and the recent research efforts to address them.  ...  We identify key characteristics of negative side effects, highlight the challenges in avoiding negative side effects, and discuss recently developed approaches, contrasting their benefits and limitations  ...  Approaches to Mitigate Negative Side Effects This section reviews the emerging approaches to mitigating the impacts of negative side effects.  ... 
arXiv:2008.12146v3 fatcat:np263lgag5ejpaywdq5o2o2lqe

Avoiding Negative Side Effects Due to Incomplete Knowledge of AI Systems

Sandhya Saisubramanian, Shlomo Zilberstein, Ece Kamar
2022 The AI Magazine  
Learning to recognize and avoid such negative side effects (NSEs) of an agent's actions is critical to improve the safety and reliability of autonomous systems.  ...  This article provides a comprehensive overview of different forms of NSEs and the recent research efforts to address them.  ...  APPROACHES TO MITIGATE NEGATIVE SIDE EFFECTS This section reviews the emerging approaches to mitigating the impacts of NSEs.  ... 
doi:10.1609/aimag.v42i4.7390 fatcat:z4bybhdzofcgxcagvmxxv2vjpy

Avoiding Negative Side Effects of Autonomous Systems in the Open World

Sandhya Saisubramanian, Ece Kamar, Shlomo Zilberstein
2022 The Journal of Artificial Intelligence Research  
Negative side effects are undesirable, unmodeled effects of agent actions on the environment.  ...  Due to the limited fidelity of the model, agent actions may produce negative side effects (NSEs) when deployed.  ...  This work performed while Sandhya was a student at the University of Massachusetts Amherst. Support for this work was provided in part by the Semiconductor Research Corporation under grant #2906.001.  ... 
doi:10.1613/jair.1.13581 fatcat:errrlznmjfg25dox47x62ejd24

Multi-criteria optimization and automated network restructuring to mitigate construction projects delays on-the-run [article]

Nina Prins, Omar Kammouh, A.R.M. Wolfert
2022 arXiv   pre-print
finding the most effective and efficient mitigation strategy.  ...  To fulfill the purpose of this paper, the Mitigation Controller (MitC) has been developed using an integrative approach of non-linear optimization techniques, probabilistic Monte Carlo simulation, and  ...  (a Dutch special contractor company) for providing the demonstrator case and for giving the possibility to verify and validate the extended MitC.  ... 
arXiv:2206.09823v1 fatcat:3xxwdnarmfcqzosi77zpfdh24a

Envisioning carbon-free land use futures for Sweden: a scenario study on conflicts and synergies between environmental policy goals

Tom van der Voorn, Åsa Svenfelt, Karin Edvardsson Björnberg, Eléonore Fauré, Rebecka Milestad
2020 Regional Environmental Change  
This requires a comprehensive multi-target backcasting approach, which combines the strengths of multicriteria analysis, nexus approaches and backcasting, for supporting a transition to zero GHG emissions  ...  In climate change mitigation, backcasting scenarios are often used for exploring options for achieving a single environmental goal, albeit at the expense of other goals.  ...  We argue that if the results and suggestions of backcasting studies are to be useful to planning and policy, we need to make sure that there are no severe side-effects for societal policy objectives.  ... 
doi:10.1007/s10113-020-01618-5 fatcat:7evmokfcpbbennfxwbdaizrcl4

Environmental Assessment of Demand-Side Management Programs

J. Turner, J. Sinclair
1996 Energy Studies Review  
From this, and the literature, a number of fA approaches are developed and evaluated. A multi-staged DSM EA framework is developed and applied to a DSM program.  ...  Demand side management (DSM) programs provide unique challenges to environmental impact assessment (EA). It is unclear whether traditional EA is warranted, or, what framework such an fA should have.  ...  The identification of mitigation methods and examples of mitigation efforts effectively presented a range of options for further study.  ... 
doi:10.15173/esr.v7i3.374 fatcat:3e75dmflfnesfg3mtouegvuewq

Sustainable development synergies and their ability to create coalitions for low-carbon transport measures

Oliver Lah
2017 Transportation Research Procedia  
This paper aims to make a contribution to understand these opportunities by highlighting the linkages between objectives, presenting case studies, facts and figures.  ...  This paper aims to make a contribution to understand these opportunities by highlighting the linkages between objectives, presenting case studies, facts and figures.  ...  The paper also explores the risks and uncertainties of some impacts of mitigation measures that may lead to trade-offs and negative side-effects.  ... 
doi:10.1016/j.trpro.2017.05.495 fatcat:kvpp4jashnbhzjcfptohwn7vfu

Fairness in Music Recommender Systems: A Stakeholder-Centered Mini Review

Karlijn Dinnissen, Christine Bauer
2022 Frontiers in Big Data  
Overall, this narrative literature review shows that the large majority of works analyze the current situation of fairness in music recommender systems, whereas only a few works propose approaches to improve  ...  As fairness is a fundamental value of human life, there is increasing pressure for these algorithmic decision-making processes to be fair as well.  ...  Embracing these attributes, they introduce a multi-objective approach to diversification that addresses fairness for users and item providers alike.  ... 
doi:10.3389/fdata.2022.913608 pmid:35937551 pmcid:PMC9353048 fatcat:adjh25phcvh7nfomgquhbr2vei

An investigation of the harmonic effects of nonlinear loads on power distributon network

T.A. Abdul-Hameed, M.F. Akorede, Y Abdulrahman, O.M. Tijani
2019 Nigerian Journal of Technology  
When present in large quantity, harmonics may cause adverse effects not only to other loads on the network but also to the power distribution network.  ...  Heavy presence of a nonlinear load or the presence of many nonlinear loads in a distribution network may significantly introduce harmonics into such network.  ...  A selective harmonics elimination (SHE) method to mitigate low order harmonics in multi-level inverters with different loads was developed in [11] .  ... 
doi:10.4314/njt.v38i2.16 fatcat:lnhbwm5wc5fqvpltm5mc6tuycy

Application of Artificial Neural Networks to Ship Detection from X-Band Kompsat-5 Imagery

Jeong-In Hwang, Sung-Ho Chae, Daeseong Kim, Hyung-Sup Jung
2017 Applied Sciences  
To remarkably reduce the negative effects in the ANN approach, we need to minimize these effects. Moreover, the Kompsat-5 SAR image has severe ghost phenomena and side-lobe effects.  ...  When ships are detected from SAR images, the most important thing is to minimize the negative effects from background noise, the side-lobe effect, wave effect, land area effect and so on.  ...  side-lobe effect, and so on.  ... 
doi:10.3390/app7090961 fatcat:amt47dku35bsnp4oiu56so6ffi

Learning from Noisy Anchors for One-stage Object Detection [article]

Hengduo Li, Zuxuan Wu, Chen Zhu, Caiming Xiong, Richard Socher, Larry S. Davis
2020 arXiv   pre-print
In this paper, we propose to mitigate noise incurred by imperfect label assignment such that the contributions of anchors are dynamically determined by a carefully constructed cleanliness score associated  ...  State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their intersection-over-union (  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes not withstanding any copyright annotation thereon.  ... 
arXiv:1912.05086v2 fatcat:ljdqle2aqzea5ijwbl23jmtvbi

Multi-Objective Real-Time Tuning of SVC Used in Electrified Traction Systems

Mohammad Hossein Bigharaz, Mehdi Amiri Dehcheshmeh, Hadi Givi, Štěpán Hubálovský
2022 Sensors  
A multi-objective optimization technique entitled non-dominated sorting genetic algorithm (NSGA-II) is used to fulfill the regarded objectives simultaneously.  ...  In this paper, a novel real-time optimization approach is presented to adjust the static VAR compensator (SVC) for the traction system to realize two objectives; current unbalance reduction and reactive  ...  Multi-Objective Optimization Generally, a multi-objective optimization problem (MOP) is considered as a set of 'P' objective functions f= (f 1 , . . .  ... 
doi:10.3390/s22041584 pmid:35214485 pmcid:PMC8877460 fatcat:45hqav2cdfcn7oh5ksmzl6pz4i

Automatic Ship Detection Using the Artificial Neural Network and Support Vector Machine from X-Band Sar Satellite Images

Jeong-In Hwang, Hyung-Sup Jung
2018 Remote Sensing  
This means that the proposed method enables the detection of most ship targets from X-band SAR images with a reduced number of false detections from negative effects.  ...  We focused on generating two optimal input data neurons that (i) strengthened ship targets and (ii) mitigated noise effects by image processing techniques, including median filtering, multi-looking, etc  ...  The two input neurons were generated to (i) emphasize ship targets from sea and (ii) mitigate the negative effects that cause false detections, e.g., speckle noise, side-lobe effects, ship wake, etc.  ... 
doi:10.3390/rs10111799 fatcat:lek3qdmnd5azrf3vc6gzviw7vi

Automatic Inference of Rule-Based Specifications of Complex In-place Model Transformations [chapter]

Timo Kehrer, Abdullah Alshanqiti, Reiko Heckel
2017 Lecture Notes in Computer Science  
In contrast to previous approaches to the same problem, our approach supports the inference of complex rule features such as negative application conditions, multi-object patterns and global invariants  ...  We follow a model transformation by-example approach to mitigate that problem: Users generate transformation rules by creating examples of transformations using standard visual editors as macro recorders  ...  As a side-effect of providing a second example, rule variables x1 to x6 are inferred denoting equalities of attribute values.  ... 
doi:10.1007/978-3-319-61473-1_7 fatcat:a2rjyxo2q5fjtaoxznintfmece
« Previous Showing results 1 — 15 out of 116,336 results